Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Barry Grossman who wrote (65161)	9/20/1998 8:44:00 PM
From: Harvey Allen	Read Replies (1) \| Respond to of 186894

Barry- Once you unwrap the flag from the presidency and begin to grade them say Washington,, Lincoln on one end and Harding on the other it gets easy. Harvey

To: Barry Grossman who wrote (65161)	9/21/1998 12:12:00 AM
From: Paul Engel	Respond to of 186894

Barry & Intel Investors - Intel isn't alone in delaying Copper Metallization Adoption. Several IC companies have evaluated copper to replace aluminum metallization and have concluded what Intel has - it is not yet necessary. A good description of their arguments follows. Paul {==========================} techweb.com September 21, 1998, Issue: 1026 Section: International Copper? Not so fast David Lammers When IBM Corp.'s stock price shot up last autumn after the company unveiled plans to use copper interconnects in PowerPC microprocessors, Motorola Inc.'s technical jaw dropped. Motorola rather off-handedly had said a similar thing a few months earlier, but didn't play the PR violin as skillfully as IBM. In fact, IBM-with so much riding on AS/400 and other servers that absolutely require fast processors-has more to gain in the near term than Motorola from an early ramp of copper. Nearly a year after IBM's headlines, a clearer view is emerging. In December, at the International Electron Devices Meeting in San Francisco, Mitsubishi Electric will argue that, for local interconnects, aluminum is faster than copper, at least for the 0.18-micron generation. The point is that each copper layer, with its superior electromigration properties, needs a silicon-nitride liner to avoid copper oxidation and diffusion and as an etch stop. For Mitsubishi, that stopper is 50 nm. Thicker wires are better than thin, and liner space obviously detracts from the metal thickness. Total propagation delay for the thicker copper-doped aluminum wires (no liner) was faster than for copper wires for the M1 to M4 layers, where the pitch is 0.53 microns, according to Mitsubishi's ULSI center. For the thicker M5 and M6 layers, where the pitch is 1.33 microns, copper wins. The crossover appears to be wires of 3 mm. Shorter than that, aluminum is faster; longer, and copper is better. Toshiba reached a similar conclusion last June. A 0.18-micron dual Damascene process using a similar copper-aluminum alloy, not copper, seemed as fast as the copper processes described. "There is only a small gain in using copper if a low-k dielectric is not integrated with it," a Toshiba engineer concluded. Intel's Mark Bohr said, "copper is real, and it will be mainstream, but the aspect ratios (of the wires) must be close to two to compete with aluminum." Intel has its 0.18-micron process ready to go, and IBM's announcement caught Intel at a point of transition. Bohr said, "certainly, at 0.13 micron, copper makes sense." Several people noted the high initial conversion cost to copper and said companies waiting a year or two will pay less for the machinery. And they may be able to combine copper with low-k dielectrics in one step. But IBM made a companywide decision to push copper. Jill Slattery, an IBM manufacturing manager, said IBM's group "felt the more complicated step was copper," and noted that the granularity of some of the low-k materials will make it challenging to do with CMP process steps. So life is not so simple, after all. Copyright r 1998 CMP Media Inc.